Title :
A geometric approach to 3D object comparison
Author :
Novotni, Marcin ; Klein, Reinhard
Author_Institution :
Comput. Graphics Group, Bonn Univ., Germany
Abstract :
Along with the development of 3D acquisition devices and methods and the increasing number of available 3D objects, new tools are necessary to automatically analyze, search and interpret these models. In this paper, we describe a novel geometric approach to 3D object comparison and analysis. To compare two objects geometrically, we first properly position and align the objects. After solving this pose estimation problem, we generate specific distance histograms that define a measure of the geometric similarity of the inspected objects. The geometric approach is very well-suited for structures with moderate variance, e.g. bones, fruits, etc. The strength of the approach is proven through the results of several tests that we performed on different data sets
Keywords :
computational geometry; object recognition; 3D acquisition devices; 3D object comparison; distance histograms; geometric similarity measure; inspection; object alignment; object positioning; pose estimation; structural variance; Bones; Computer graphics; Computer vision; Content based retrieval; Histograms; Position measurement; Principal component analysis; Solid modeling; Testing; Web sites;
Conference_Titel :
Shape Modeling and Applications, SMI 2001 International Conference on.
Conference_Location :
Genova
Print_ISBN :
0-7695-0853-7
DOI :
10.1109/SMA.2001.923387